Article
Twenty-six hours to three: a global drought index at four kilometres
Seventy-six years of monthly water balance for every land pixel on Earth at four kilometre resolution used to take twenty-six hours to compute. It now takes under three.
Article
Seventy-six years of monthly water balance for every land pixel on Earth at four kilometre resolution used to take twenty-six hours to compute. It now takes under three.
Article
The closing post. What the horizon looks like once the warming trend is taken out, how little independent information nine hundred months contains, and the three things this project got wrong and had to fix.
Article
Evaluating the fitted relationship at a forecast beyond anything used to fit it. The chance of a record season runs four to five times higher than chance, about half of that is the warming trend, most of the world has no business being asked, and
Article
The strongest matching seasons on record, composited. Rainfall at 61 per cent of normal across the Maritime Continent, five analogues out of five agreeing, and a correction that has to be applied before any of it can be read.
Article
Not a forecast and not a composite: the observed state of both halves of the water balance through August 2026. The Sahel is at SPI-3 of -2.39, half the Maritime Continent is already in a dry spell, and a quarter of the world’s land is dry on one
Article
Anak Krakatau erupted at midnight and the ground observers could not see the ash column. A satellite photographs the whole hemisphere every ten minutes, in sixteen colours. Here is what each of them shows.
Article
The hypothesis this project was built to test was that El Nino makes some places wet and thirsty at once. On land where both responses are solid, that combination is 0.3 per cent. What is there instead is worse.
Article
Run theory on both halves of the water balance at four accumulation lengths. Globally El Niño changes almost nothing. Over the Maritime Continent it changes the share of months that come in below the dry mark by nineteen points, and it does it by
Article
Drought is what arrives minus what leaves, and almost everything published measures only the first. This post builds the second, finds that a thirty-year baseline cannot express two of its own eleven categories, and then sets out what all of it is
Article
Rebuilding the thesis in Latin Modern took about thirty kilobytes of shell script, and most of it was not about the font. It was about making sure nobody could mistake the result for the signed copy.
Article
Seventy-six years of record contains forty-four ENSO events. Twenty-two are El Niño, eight of those are strong, and every single one of the eight is the same flavour. That is the whole evidence base.
Article
In February 2026 the Climate Prediction Center changed the index it uses to classify El Niño. The two indices disagree by 0.41 degrees right now, and the choice moves the map on 95 per cent of land.
Article
In six months the Pacific swung from La Niña to El Niño conditions, and the official outlook has it climbing past anything in the record. For four consecutive seasons the central forecast sits above every value observed since 1950.
Article
The graduation letter arrived today. What the two years actually cost, and the slides from the three points where the work had to be defended out loud.
Article
I am not fluent in Word. Two hundred pages, a hundred and fifty figures and a moving bibliography made that a problem, and switching was the best decision I made.
Article
Passed, revision required. The substantive feedback was not that anything was wrong, but that the whole thing started in the wrong place.
Article
Three reviewers, and the first one found something structural: my tables were scored against the thing I had calibrated to, which made the simplest method look best.
Article
The paper is out in Remote Sensing. What it argues, where everything lives, and the one number in it I would most like people to take away.
Article
The code and the data have DOIs. The paper does not yet. Which turns out to be the clearest way to explain why one identifier was never going to be enough.
Article
NetCDF is the right home for gridded climate data and the wrong shape for combining it with anything else. Zonal statistics onto an H3 grid turns a raster into a table you can join on, four million hexagons at a time.